CN101308202A - Parallel collection image reconstruction method and device - Google Patents

Parallel collection image reconstruction method and device Download PDF

Info

Publication number
CN101308202A
CN101308202A CNA2007101075109A CN200710107510A CN101308202A CN 101308202 A CN101308202 A CN 101308202A CN A2007101075109 A CNA2007101075109 A CN A2007101075109A CN 200710107510 A CN200710107510 A CN 200710107510A CN 101308202 A CN101308202 A CN 101308202A
Authority
CN
China
Prior art keywords
matrix
sampling
reconstruction
data
image
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CNA2007101075109A
Other languages
Chinese (zh)
Other versions
CN101308202B (en
Inventor
汪坚敏
张必达
傅彩霞
张强
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Siemens AG
Original Assignee
Siemens Ltd China
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Siemens Ltd China filed Critical Siemens Ltd China
Priority to CN2007101075109A priority Critical patent/CN101308202B/en
Priority to US12/152,573 priority patent/US7576539B2/en
Publication of CN101308202A publication Critical patent/CN101308202A/en
Application granted granted Critical
Publication of CN101308202B publication Critical patent/CN101308202B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00—Arrangements or instruments for measuring magnetic variables
    • G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
    • G01R33/44—Arrangements or instruments for measuring magnetic variables involving magnetic resonance using nuclear magnetic resonance [NMR]
    • G01R33/48—NMR imaging systems
    • G01R33/54—Signal processing systems, e.g. using pulse sequences ; Generation or control of pulse sequences; Operator console
    • G01R33/56—Image enhancement or correction, e.g. subtraction or averaging techniques, e.g. improvement of signal-to-noise ratio and resolution
    • G01R33/561—Image enhancement or correction, e.g. subtraction or averaging techniques, e.g. improvement of signal-to-noise ratio and resolution by reduction of the scanning time, i.e. fast acquiring systems, e.g. using echo-planar pulse sequences
    • G01R33/5611—Parallel magnetic resonance imaging, e.g. sensitivity encoding [SENSE], simultaneous acquisition of spatial harmonics [SMASH], unaliasing by Fourier encoding of the overlaps using the temporal dimension [UNFOLD], k-t-broad-use linear acquisition speed-up technique [k-t-BLAST], k-t-SENSE

Landscapes

  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Radiology & Medical Imaging (AREA)
  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • High Energy & Nuclear Physics (AREA)
  • Condensed Matter Physics & Semiconductors (AREA)
  • General Physics & Mathematics (AREA)
  • Magnetic Resonance Imaging Apparatus (AREA)

Abstract

The invention discloses a method for reconstructing a parallel-acquired image, comprising: generating reconstruction data by combining uniformly under-sampled data and low-frequency fully-sampled data in magnetic resonance imaging (MRI) K-space according to a hybrid sampling mode; calculating the sensitivity distribution of a coil according to the low-frequency fully-sampled data; and reconstructing an image according to the reconstruction data, the sensitivity distribution of the coil and the hybrid sampling mode. The invention also discloses an apparatus for reconstructing the parallel-acquired image. The method and the apparatus are adopted to generate reconstruction data by combining uniformly under-sampled data and low-frequency fully-sampled data in K-space according to the hybrid sampling mode and take the hybrid sampling mode into account during the image reconstruction, and the signal to noise ratio of the reconstructed image is effectively improved by using the reconstruction data combined with the low-frequency fully-sampled data in reconstructing the image since the low-frequency fully-sampled data contains more useful information.

Description

The method and apparatus of parallel collection image reconstruction
Technical field
The present invention relates to magnetic resonance imaging (MRI, Magnetic Resonance Imaging) technology, particularly relate to the method and apparatus of a kind of parallel acquisition (PA, Parallel Acquisition) image reconstruction.
Background technology
In the MRI technology, the speed of imaging is unusual important parameters.Early stage inspection usually needs to spend several hours, and owing to the improvement of technology aspect field intensity, gradient hardware and pulse train, the speed of imaging has had a bigger lifting subsequently.But, continuous radio frequency (the RF of field gradient Fast transforms and high density, Radio Frequency) the pulse meeting brings the unaffordable specific absorption rate (SAR of the Human Physiology limit, Specific Absorption Rate) and the pyrogenicity amount of organ-tissue, therefore, the lifting of image taking speed has run into bottleneck.
Subsequently, the researchist finds that by using complicated computerized image reconstruction algorithm and the coil array that matches, the image taking speed of magnetic resonance can be promoted greatly, and this kind technology is commonly called parallel imaging (parallel imaging) technology.Parallel collection image reconstruction is a kind of image reconstruction (reconstruction) technology that is used for parallel acquisition, it utilizes spatial sensitivity (sensitivity) difference of phased-array coil to carry out space encoding, and gather simultaneously with phased-array coil, obtain doubly even higher image taking speed than the fast 2-6 of conventional magnetic resonance imaging.Adopt parallel imaging technique, magnetic resonance imaging system is proposed new requirement, as a plurality of receiving cables of needs, complex array coil and coil sensitivity calibration, with special data processing and image rebuilding method or the like.
Parallel imaging can improve image acquisition speed, and the raising of image taking speed realizes by reducing the K space filling factor.If being lower than the restriction of sampling thheorem, the K space filling factor can cause directly utilizing the image after Fourier is rebuild pseudo-shadow to occur.General magnetic resonance imaging is to obtain image by the frequency domain information of gathering object by Fourier transform.According to sampling thheorem, be inversely proportional to the sampling interval of frequency domain in the repetition period of image area object.If the space repetition period of image, the image after the reconstruction will overlap less than the size of image itself, this phenomenon is called aliasing in signal Processing.If equal the inverse of phase-encoding direction FOV in the sampling interval of phase-encoding direction, image just just can not overlap, claim this moment the K space be sampled as full sampling.The sampling of sampling interval during greater than 1/FOV is called owes sampling, that is to say that the K spatial information that collects is not enough to rebuild complete image, owes to sample to cause the overlapping of image; Otherwise be over-sampling, also can not cause the doubling of the image.The synoptic diagram that Fig. 1 shows full sampling and owes to sample, wherein left figure is full sampling, right figure owes sampling.In Fig. 1, vertical direction is a phase-encoding direction, and left side figure is full sampling, and the sampling interval on phase-encoding direction is Δ Ky; Right figure be 1/2 owe the sampling, the sampling interval on phase-encoding direction is 2 Δ Ky.
In magnetic resonance imaging, the magnetic resonance signal of tissue is by the receiving coil collection.Fig. 2 is the image that phase-encoding direction shown in Figure 1 is completely sampled and obtained when owing to sample.Among Fig. 2, vertical direction is a phase directional, and horizontal direction is a frequency direction.As shown in Figure 2, the image that left figure obtains when being full the sampling, sampling interval is Δ Ky; Right figure is 1/2 image of owing to sample and obtaining, and sampling interval is 2 Δ Ky.As shown at right, if (Coil1 and Coil2) gathers the sample among the figure respectively with two adjacent surface coils, and the sampling rate of phase-encoding direction is 2/FOV, will obtain having two width of cloth images of overlapping pseudo-shadow after directly rebuilding, the visual field of every width of cloth image is FOV/2, but the content that all comprises another width of cloth figure among every width of cloth figure, it is the contribution that the value of each pixel among the right figure comes from two volume elements among the former figure, for instance, the signal intensity of pixel 1 comes from two volume elements 1 among the left figure among the right figure, 2 contribution, equally, the signal intensity of pixel 2 also comes from two volume elements 2 among the left figure among the right figure, 1 contribution.From right figure, be not difficult to observe a little less than the signal intensity of signal intensity than original image of pseudo-shadow, this be since the sample spin signal that coil collects by the sensitivity function weighting of coil, and the amplitude of the sensitivity function of surface coils is far away regional less of distance coil.The aliasing that sampling causes depends on the filling mode in K space, the just track in K space.Pseudo-shadow above-mentioned is the pseudo-shadow that uniform grid is owed to sample and caused, for helical trajectory, the pseudo-shadow that causes of owing to sample will be very irregular.
For multi-thread circle collection, though the K spatial information that each coil collects is not enough, the difference that can utilize different coils to collect signal obtains a complete image through handling.The reconstruction algorithm of the pseudo-shadow of parallel imaging anti-aliasing can be divided into two classes haply: space harmonics parallel acquisition (SMASH, SiMultaneous Acquisition of Spatial Harmonics) and sensitivity encoding parallel acquisition technique (SENSE, SENSitivity Endoding parallel acquisitiontechnique).Wherein, the SMASH method is a kind of method of utilizing the sensitivity function composition space harmonics of each passage coil and carrying out auxiliaring coding.The sensitivity function of general coil is slowly to change in the space, and can be similar to and think gauss of distribution function, so can utilize the linear combination of the sensitivity function of each passage coil to constitute the space harmonics of certain frequency.And utilize the space harmonics function to gather out not have the phase encoding of actual acquisition line.
The scheme of handling at frequency domain with SMASH is different, and the SENSE method is eliminated the pseudo-shadow of owing to sample and causing by the method for finding the solution system of linear equations at image area.Because the eclipsing effects that causes of image space cycle, for instance, when sampling rate for full sampling 1/2 the time, directly utilize of the contribution of the value of each pixel in the image that Fourier transform obtains from two volume elements in the original image.Can explain with formula (1):
s 1＝p 11m 1+p 12m 2(1)
Formula (1) can cooperate Fig. 2 to illustrate: the value of right figure mid point 1 by the magnetization m1 of volume elements 1 among the left figure multiply by coil 1 the magnetization m2 of the sensitivity P11 at point 1 place and volume elements 2 multiply by coil 1 point 2 places sensitivity P12's and decide.But P11 and P12 are measuring amount, want to obtain m1 and m2, need two equations at least, i.e. the data that need two absolute coils to measure at least.With matrix representation be exactly,
s 1 s 2 = p 11 , p 21 p 12 , p 22 · m 1 m 2 - - - ( 2 )
Wherein S2 is the value of right figure mid point 2, and the definition of P21 and P22 is the same.Formula (2) can further be expressed as:
S＝PM (3)
Wherein S is the vector of Nc * 1, and Nc is a port number.M is the vector of Np * 1, and Np is for owing sampling rate.P inverted just obtain original image.For port number greater than the situation of owing sampling rate, the Nc＞Np in the equation (3), this is an overdetermination system of linear equations, promptly the number of known conditions is greater than the number of unknown number.Generally ask the plus sige of P to reach optimum against making to separate.
Wherein, matrix W is called as weighting matrix, and superscript H represents conjugate transpose.
In radio frequency electromagnetic field, human body is regarded as load, and the sensitivity function of coil can change when measuring different human bodies to some extent, and these variations are enough to influence quality of reconstructed images, so the sensitivity of coil generally need be obtained in real time.Completely sample in the k central zone of space during SENSE method image data, owe sampling in outer peripheral areas.So original K spatial data is divided into two parts: evenly owe the data that the data of sampling and low frequency are completely sampled.Evenly owe the data of the sampling image that generates aliasing, and the data that low frequency is completely sampled are with generating fuzzy tissue image and then obtaining real-time coil sensitivity profiles and weighting matrix, to utilize data aliased image that generates and the weighting matrix that utilizes the full sampled data of low frequency to obtain of evenly owing to sample to synthesize at last, obtain high-resolution no aliased image.Here, the full sampled data of low frequency that is used to obtain coil sensitivity profiles and weighting matrix is called reference data, and the phase encoding line that K space medium and low frequency is completely sampled is called reference line.
The advantage of utilizing hyperchannel coil parallel acquisition to improve image taking speed is conspicuous.If under the identical sampling time, utilize parallel acquisition technique also can mention the resolution of image.Parallel acquisition also can attach some benefits, such as alleviating owing to the image artifacts that goes resonance (off-resonance) to cause.Simultaneously, because the raising of picking rate, parallel acquisition technique can also alleviate motion artifacts.
Yet the realization parallel acquisition need be paid certain cost.At first and since actual acquisition to data reduced, though utilize the parallel acquisition reconstruction technique can reconstruct the image of no aliasing, the signal to noise ratio (S/N ratio) of image is the full signal noise ratio (snr) of image of sampling and obtaining under the same hardware condition
Figure A20071010751000073
N wherein fBe speedup factor, therefore, speedup factor is big more, and picture quality reduces big more, and promptly signal to noise ratio (S/N ratio) (SNR, Signal-to-Noise Ratio) reduces big more.
Summary of the invention
One object of the present invention is to propose a kind of method of parallel collection image reconstruction, improves the SNR that parallel acquisition is rebuild the back image.
Another object of the present invention is to propose a kind of device of parallel collection image reconstruction, improve the SNR that parallel acquisition is rebuild the back image.
To achieve these goals, technical scheme of the present invention comprises:
A kind of method of parallel collection image reconstruction, it comprises: evenly owe the data reconstruction that sampled data and the full sampled data of low frequency combine according to mixing K space that sampling pattern generates magnetic resonance MR imaging; Calculate coil sensitivity profiles according to the full sampled data of described low frequency; According to described data reconstruction, described coil sensitivity profiles and described mixing sampling pattern reconstructed image.
Wherein, describedly evenly owe the data reconstruction that sampled data and the full sampled data of low frequency combine and comprise: generate the data reconstruction matrix according to mixing K space that sampling pattern generates magnetic resonance MR imaging, in described data reconstruction matrix, evenly owe sampled data and the full sampled data of low frequency for the K space of actual acquisition, fill with the actual acquisition value, for the data of not gathering, fill with 0.
Wherein, the line number of described data reconstruction matrix is acquisition channel number, phase encoding number and the product of reading the direction sampling number, and columns is 1.
Wherein, described data computation coil sensitivity profiles of completely sampling according to low frequency comprises: calculate the coil sensitivity profiles matrix according to the full sampled data of low frequency, and arrangement and the dimension of determining described sensitivity profile matrix according to the arrangement and the dimension of described data reconstruction matrix.
Wherein, described according to data reconstruction, coil sensitivity profiles and mix the sampling pattern reconstructed image and comprise: the mathematic(al) representation of determining described mixing sampling pattern; Mathematic(al) representation reconstructed image according to described data reconstruction, described coil sensitivity profiles matrix and described mixing sampling pattern.
Wherein, the mathematic(al) representation that described generation mixes sampling pattern comprises: generate and mix sampling matrix, on the diagonal line of described mixing sampling matrix, fill at interval with 0 and 1 according to owing sampling rate, and the relevant position of completely sampling in the centre fills with 1, and other positions except that diagonal line are all with 0 filling.
Wherein, describedly comprise: arrangement and the dimension of determining the inverse Fourier transform matrix according to the arrangement and the dimension of described data reconstruction matrix according to data reconstruction, coil sensitivity profiles matrix and the mathematic(al) representation reconstructed image that mixes sampling pattern; Utilize described inverse Fourier transform matrix, described mixing sampling matrix and coil sensitivity profiles matrix computations to mix the parallel imaging sampling matrix of sampling pattern: A=M UndersamplingM FourierM Sensitivity, wherein, M FourierExpression inverse Fourier transform matrix, M SensitivityExpression coil sensitivity profiles matrix, M UndersamplingExpression mixes sampling matrix, and A represents to mix the parallel imaging sampling matrix of sampling pattern: contrary as reconstruction matrix with the parallel imaging sampling matrix of described mixing sampling pattern; Utilize described reconstruction matrix and described data reconstruction matrix reconstructed image: ima=(A ' A) -1A ' V Rawdata, wherein, ima represents reconstructed image matrix, V RawdataExpression data reconstruction matrix.
Wherein, parallel imaging sampling matrix and data reconstruction matrix reconstructed image that described utilization mixes sampling pattern comprise: with described data reconstruction matrix decomposition is N vector, each vector comprises M element, wherein M is the product of acquisition channel number and phase encoding line number, and N is for reading the direction sampling number; Respectively a described N vector is rebuild, obtained N image vector; With described N image vector combination, obtain the reconstructed image matrix.
A kind of device of parallel collection image reconstruction, it comprises: the data reconstruction generation unit is used for evenly owing the data reconstruction that sampled data and the full sampled data of low frequency combine according to mixing K space that sampling pattern generates magnetic resonance MR imaging; The sensitivity profile computing unit is used for calculating coil sensitivity profiles according to the full sampled data of described low frequency; Image reconstruction unit is used for according to described data reconstruction, described coil sensitivity profiles and described mixing sampling pattern reconstructed image.
Wherein, described data reconstruction generation unit generates the data reconstruction matrix, in described data reconstruction matrix, evenly owes sampled data and the full sampled data of low frequency for the K space of actual acquisition, fills with the actual acquisition value, for the data of not gathering, fills with 0.
Wherein, described sensitivity profile computing unit calculates according to the full sampled data of described low frequency and generates the coil sensitivity profiles matrix, and arrangement and the dimension of determining described coil sensitivity profiles matrix according to the arrangement and the dimension of described data reconstruction matrix.
Wherein, described image reconstruction unit comprises: mix the sampling matrix generation module, be used for generating the mixing sampling matrix; Inverse Fourier transform matrix generation module is used for arrangement and the dimension of determining the inverse Fourier transform matrix according to the arrangement and the dimension of described data reconstruction matrix, and generates described inverse Fourier transform matrix; The reconstruction matrix generation module, be used to utilize described inverse Fourier transform matrix, described sensitivity profile matrix and described mixing sampling matrix to generate the parallel imaging sampling matrix that mixes sampling pattern, and contrary as reconstruction matrix with the parallel imaging sampling matrix of described mixing sampling pattern; The image reconstruction computing module is used to utilize described reconstruction matrix and described data reconstruction matrix reconstructed image.
Wherein, described mixing sampling matrix generation module is when generating the mixing sampling matrix, fill at interval with 0 and 1 according to owing sampling rate on the diagonal line of described mixing sampling matrix, and the relevant position of completely sampling in the centre fills with 1, other positions except that diagonal line are all with 0 filling.
From above technical scheme as can be seen, in the present invention, combine as the parallel acquisition data that are used to rebuild with the full sampled data of low frequency evenly owing sampled data in the K space, because being positioned at K space center partial data comprises more useful information, therefore the data that K space low frequency is completely sampled can comprise more useful information.So, utilize the parallel acquisition data that combine the full sampled data of low frequency to carry out image reconstruction, more useful information can be provided, thereby effectively improve the SNR of reconstructed image.
Description of drawings
To make clearer above-mentioned and other feature and advantage of the present invention of those of ordinary skill in the art by describing the preferred embodiments of the present invention in detail with reference to accompanying drawing below, identical label is represented identical parts, in the accompanying drawing:
Fig. 1 is the synoptic diagram that K phase encode direction is completely sampled and owed to sample in the MRI scanning;
Fig. 2 is the image that phase-encoding direction shown in Figure 1 is completely sampled and obtained when owing to sample;
Fig. 3 is the method flow diagram of parallel collection image reconstruction according to an embodiment of the invention;
Fig. 4 (a) shows and evenly owes the phase encoding of sampling line;
Fig. 4 (b) shows the phase encoding line that low frequency is completely sampled;
Fig. 4 (c) shows the combination of the phase encoding line of uniform sampling and the phase encoding line that low frequency is completely sampled;
Fig. 5 is the synoptic diagram that mixes sampling matrix according to an embodiment of the invention;
Fig. 6 is the device process flow diagram of parallel collection image reconstruction according to an embodiment of the invention;
Fig. 7 be consider under the situation of reference line with the situation of not considering reference line under the SNR comparison diagram of reconstructed image.
Embodiment
In order to make purpose of the present invention, technical scheme and advantage clearer,, the present invention is further elaborated below in conjunction with drawings and Examples.Should be appreciated that specific embodiment described herein only in order to explanation the present invention, and be not used in qualification the present invention.
According to the present invention, expire sampled data and evenly owe the data reconstruction that sampled data combines according to the low frequency that mixes in the sampling pattern generation K space, and when image reconstruction, consider to mix sampling pattern.
Below by a specific embodiment method of the present invention is described in detail.
Fig. 3 is the method flow diagram of parallel collection image reconstruction according to an embodiment of the invention.As shown in Figure 3, in the present embodiment, the method for parallel collection image reconstruction comprises the steps: at least
Step 301: generate the full sampled data of K space low frequency and evenly owe the data reconstruction that sampled data combines according to mixing sampling pattern.
The K spatial data gathered comprise that collection evenly owes the data and the reference data of sampling.Reference data is the data that the low frequency part in K space is completely sampled and obtained.
Fig. 4 (a), Fig. 4 (b) and Fig. 4 (c) will evenly owe sampled data and the low frequency synoptic diagram that full sampled data combines.Wherein Fig. 4 (a) shows and evenly owes the phase encoding of sampling line, and Fig. 4 (b) shows the phase encoding line that low frequency is completely sampled, and Fig. 4 (c) shows the combination of the phase encoding line of uniform sampling and the phase encoding line that low frequency is completely sampled.Suppose that the speedup factor of owing to sample in the K space is 2, promptly sampling rate is 1/2, for instance, if full sampling needs to gather 17 phase encoding lines on the phase-encoding direction, is expressed as k respectively 8To k -8, sampling rate is that 9 phase encoding lines need are gathered in 1/2 the sampling of evenly owing, and illustrates as solid line among Fig. 4 (a), these 9 phase encoding lines are expressed as [k respectively 8, k 6, k 4, k 2, k 0, k -2, k -4, k -6, k -8]; Reference data takies 5 phase encoding lines of the most close K space center, and promptly 5 reference lines illustrate as solid line among Fig. 4 (b), are expressed as [k respectively 2, k 1, k 0, k -1, k -2]; The data and the reference data combination of uniform sampling are obtained data reconstruction, consequently, union is got in the set of phase encoding line and the reference line set of evenly owing to sample, and the data that will be somebody's turn to do and concentrate are as data reconstruction, illustrate as solid line among Fig. 4 (c), the phase encoding line that should and concentrate is expressed as [k respectively 8, k 6, k 4, k 2, k 1, k 0, k -1, k -2, k -4, k -6, k -8].
With the matrix representation of the data reconstruction of the data of evenly owing to sample and reference data be combined into is V Rawdata, its length calculation formula is:
V RawdataLength=nChnPEnRO (5)
Wherein, nCh is the acquisition channel number, and nPE is a phase encoding line number, and nRO reads the direction sampling number, at V RawdataIn the matrix, the data on the phase encoding line of actual acquisition are constant, and the data on the phase encoding line of owing to adopt are 0.Preferably, the line number of data reconstruction matrix is acquisition channel number, phase encoding line number and the product of reading the direction sampling number, and columns is 1.
Step 302: calculate the coil sensitivity profiles matrix M according to the full sampled data of low frequency Sensitivity
M SensitivityArrangement and dimension according to V RawdataArrangement and dimension determine.
Step 303: determine to mix the mathematic(al) representation of sampling pattern, preferably, determine to mix sampling matrix M Undersampling
Mix sampling matrix M UndersamplingExample as shown in Figure 5, as can be seen from Figure 5, at M UndersamplingDiagonal line on, evenly owe the position of sampling and fill at interval with 1 and 0 according to sampling rate, fill with 1 the relevant position of middle full sampling, 0 filling is all used in the position beyond the diagonal line.
Step 304: according to the coil sensitivity profiles matrix M Sensitivity, the data reconstruction matrix V RawdataWith mixing sampling matrix M UndersamplingReconstructed image.
At first need according to V RawdataThe arrangement of matrix and dimension are determined the inverse Fourier transform matrix M FourierArrangement and dimension;
Then, generate the parallel imaging sampling matrix that mixes sampling pattern according to formula (6):
A＝M Undersampling·M Fourier·M Sensitivity (6)
Wherein, M FourierExpression inverse Fourier transform matrix, M SensitivityExpression coil sensitivity profiles matrix, M UndersamplingExpression mixes sampling matrix, and A represents to mix the parallel imaging sampling matrix of sampling pattern.
Then, will mix parallel imaging sampling matrix A contrary of sampling pattern as reconstruction matrix, and according to formula (7) reconstructed image:
ima＝(A′·A) -1·A′·V rawdata(7)
Wherein, ima represents the reconstructed image matrix.
Preferably, in actual applications, can be with the data reconstruction matrix V RawdataBe decomposed into N vector, being about to the data reconstruction matrix decomposition is N vector, and each vector comprises M element, and wherein M is the product of acquisition channel number and phase encoding line number, and N is for reading the direction sampling number; Respectively a described N vector is rebuild, obtained N image vector; Then, described N image vector combination obtained the reconstructed image matrix.
Fig. 6 shows the structural representation of the parallel collection image reconstruction device of one specific embodiment according to the present invention.As shown in Figure 6, in the present embodiment, the parallel collection image reconstruction device comprises data reconstruction generation unit 601, sensitivity profile computing unit 602 and image reconstruction unit 603 at least.
Data reconstruction generation unit 601 evenly owes sampled data according to the K space of mixing sampling pattern generation MR imaging and low frequency is expired the data reconstruction that sampled data combines.
Sensitivity profile computing unit 602 calculates coil sensitivity profiles according to the full sampled data of low frequency.
Image reconstruction unit 603 is according to data reconstruction, coil sensitivity profiles and mixing sampling pattern reconstructed image.
Preferably, data reconstruction generation unit 601 generates the data reconstruction matrix, in the data reconstruction matrix, evenly owes sampled data and the full sampled data of low frequency for the K space of actual acquisition, fills with the actual acquisition value, for the data of not gathering, fills with 0.
Preferably, sensitivity profile computing unit 602 calculates according to the full sampled data of low frequency and generates the coil sensitivity profiles matrix, and arrangement and the dimension of determining the coil sensitivity profiles matrix according to the arrangement and the dimension of data reconstruction matrix.
Further, image reconstruction unit 603 comprises mixing sampling matrix generation module 6031, inverse Fourier transform matrix generation module 6032, reconstruction matrix generation module 6033 and image reconstruction computing module 6043.
Mix sampling matrix generation module 6031 and generate the mixing sampling matrix;
Arrangement and dimension that inverse Fourier transform matrix generation module 6032 is determined the inverse Fourier transform matrix according to the arrangement and the dimension of data reconstruction matrix, and generate the inverse Fourier transform matrix;
Reconstruction matrix generation module 6033 utilizes inverse Fourier transform matrix, sensitivity profile matrix and mixes sampling matrix and generates the parallel imaging sampling matrix that mixes sampling pattern, and will mix sampling pattern the parallel imaging sampling matrix against as reconstruction matrix;
Image reconstruction computing module 6043 utilizes reconstruction matrix and data reconstruction matrix reconstructed image.
Preferably, mix sampling matrix generation module 6031 when generating the mixing sampling matrix, fill at interval with 0 and 1 according to owing sampling rate on the diagonal line of described mixing sampling matrix, and the relevant position of completely sampling in the centre fills with 1, other positions except that diagonal line are all with 0 filling.
As mentioned above, in the present invention, combine with the full sampled data of low frequency evenly owing sampled data in the K space, as the parallel acquisition data that are used to rebuild, because being positioned at K space center partial data comprises more useful information, therefore the data that K space low frequency is completely sampled can comprise more useful information.So the parallel acquisition data that combine the full sampled data of low frequency can provide more useful information, thereby effectively improve the SNR of reconstructed image behind Fourier transform.Be appreciated that the variation with the ratio of the quantity of total phase encoding line changes SNR improvement situation according to reference line, in general, the ratio that reference line accounts for total phase encoding line is big more, and SNR is high more.
Fig. 7 shows the SNR that ratio that reference line accounts for total phase encoding line is a reconstructed image under the SNR of reconstructed image under 10% the situation and the situation of not considering reference line.In Fig. 7, the ratio that the curve representation reference line of top accounts for total phase encoding line is the SNR of reconstructed image under 10% the situation, and the curve representation of below is not considered the SNR of reconstructed image under the situation of reference line.As can be seen from Figure 7, compare with the SNR of reconstructed image under the situation of not considering reference line, the ratio that reference line accounts for total phase encoding line is under 10% the situation, and SNR improves and is about 2%.
The above only is preferred embodiment of the present invention, not in order to restriction the present invention, all any modifications of being done within the spirit and principles in the present invention, is equal to and replaces and improvement etc., all should be included within protection scope of the present invention.

Claims (13)

1, a kind of method of parallel collection image reconstruction is characterized in that, comprising:
Evenly owe the data reconstruction that sampled data and the full sampled data of low frequency combine according to mixing K space that sampling pattern generates magnetic resonance MR imaging;
Calculate coil sensitivity profiles according to the full sampled data of described low frequency;
According to described data reconstruction, described coil sensitivity profiles and described mixing sampling pattern reconstructed image.
2, method according to claim 1 is characterized in that, describedly evenly owes the data reconstruction that sampled data and the full sampled data of low frequency combine and comprises according to mixing K space that sampling pattern generates magnetic resonance MR imaging:
Generate the data reconstruction matrix, in described data reconstruction matrix, evenly owe sampled data and the full sampled data of low frequency, fill,, fill with 0 for the data of not gathering with the actual acquisition value for the K space of actual acquisition.
3, method according to claim 2 is characterized in that, the line number of described data reconstruction matrix is acquisition channel number, phase encoding number and the product of reading the direction sampling number, and columns is 1.
4, method according to claim 2 is characterized in that, described data computation coil sensitivity profiles of completely sampling according to low frequency comprises:
Calculate the coil sensitivity profiles matrix according to the full sampled data of low frequency, and arrangement and the dimension of determining described sensitivity profile matrix according to the arrangement and the dimension of described data reconstruction matrix.
5, method according to claim 3 is characterized in that, and is described according to data reconstruction, coil sensitivity profiles and mix the sampling pattern reconstructed image and comprise:
Determine the mathematic(al) representation of described mixing sampling pattern;
Mathematic(al) representation reconstructed image according to described data reconstruction, described coil sensitivity profiles matrix and described mixing sampling pattern.
6, method according to claim 5 is characterized in that, the mathematic(al) representation that described generation mixes sampling pattern comprises:
Generate to mix sampling matrix, fill at interval with 0 and 1 according to owing sampling rate on the diagonal line of described mixing sampling matrix, and the relevant position of completely sampling in the centre is with 1 filling, other positions except that diagonal line are all with 0 filling.
7, method according to claim 6 is characterized in that, described mathematic(al) representation reconstructed image according to data reconstruction, coil sensitivity profiles matrix and mixing sampling pattern comprises:
Arrangement and the dimension of determining the inverse Fourier transform matrix according to the arrangement and the dimension of described data reconstruction matrix;
Utilize described inverse Fourier transform matrix, described mixing sampling matrix and coil sensitivity profiles matrix computations to mix the parallel imaging sampling matrix of sampling pattern:
A＝M Undersampling·M Fourier·M Sensitivity
Wherein, M FourierExpression inverse Fourier transform matrix, M SensitivityExpression coil sensitivity profiles matrix, M UndersamplingExpression mixes sampling matrix, and A represents to mix the parallel imaging sampling matrix of sampling pattern:
Contrary with the parallel imaging sampling matrix of described mixing sampling pattern as reconstruction matrix;
Utilize described reconstruction matrix and described data reconstruction matrix reconstructed image:
ima＝(A′·A) -1·A′·V rawdata
Wherein, ima represents reconstructed image matrix, V RawdataExpression data reconstruction matrix.
8, method according to claim 7 is characterized in that, parallel imaging sampling matrix and data reconstruction matrix reconstructed image that described utilization mixes sampling pattern comprise:
With described data reconstruction matrix decomposition is N vector, and each vector comprises M element, and wherein M is the product of acquisition channel number and phase encoding number, and N is for reading the direction sampling number;
Respectively a described N vector is rebuild, obtained N image vector;
With described N image vector combination, obtain the reconstructed image matrix.
9, a kind of device of parallel collection image reconstruction is characterized in that, comprising:
Data reconstruction generation unit (601) is used for evenly owing the data reconstruction that sampled data and the full sampled data of low frequency combine according to mixing K space that sampling pattern generates magnetic resonance MR imaging;
Sensitivity profile computing unit (602) is used for calculating coil sensitivity profiles according to the full sampled data of described low frequency;
Image reconstruction unit (603) is used for according to described data reconstruction, described coil sensitivity profiles and described mixing sampling pattern reconstructed image.
10, device according to claim 9, it is characterized in that, described data reconstruction generation unit (601) generates the data reconstruction matrix, in described data reconstruction matrix, evenly owe sampled data and the full sampled data of low frequency for the K space of actual acquisition, fill with the actual acquisition value,, fill with 0 for the data of not gathering.
11, device according to claim 10, it is characterized in that, described sensitivity profile computing unit (602) calculates according to the full sampled data of described low frequency and generates the coil sensitivity profiles matrix, and arrangement and the dimension of determining described coil sensitivity profiles matrix according to the arrangement and the dimension of described data reconstruction matrix.
12, device according to claim 11 is characterized in that, described image reconstruction unit (603) comprising: mix sampling matrix generation module (6031), be used for generating the mixing sampling matrix;
Inverse Fourier transform matrix generation module (6032) is used for arrangement and the dimension of determining the inverse Fourier transform matrix according to the arrangement and the dimension of described data reconstruction matrix, and generates described inverse Fourier transform matrix;
Reconstruction matrix generation module (6033), be used to utilize described inverse Fourier transform matrix, described sensitivity profile matrix and described mixing sampling matrix to generate the parallel imaging sampling matrix that mixes sampling pattern, and contrary as reconstruction matrix with the parallel imaging sampling matrix of described mixing sampling pattern;
Image reconstruction computing module (6043) is used to utilize described reconstruction matrix and described data reconstruction matrix reconstructed image.
13, device according to claim 12, it is characterized in that, described mixing sampling matrix generation module (6031) is when generating the mixing sampling matrix, on the diagonal line of described mixing sampling matrix, fill at interval with 0 and 1 according to owing sampling rate, and the relevant position of completely sampling in the centre fills with 1, and other positions except that diagonal line are all with 0 filling.
CN2007101075109A 2007-05-17 2007-05-17 Parallel collection image reconstruction method and device Expired - Fee Related CN101308202B (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN2007101075109A CN101308202B (en) 2007-05-17 2007-05-17 Parallel collection image reconstruction method and device
US12/152,573 US7576539B2 (en) 2007-05-17 2008-05-15 Method and apparatus for reconstructing parallel-acquired image

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2007101075109A CN101308202B (en) 2007-05-17 2007-05-17 Parallel collection image reconstruction method and device

Publications (2)

Publication Number Publication Date
CN101308202A true CN101308202A (en) 2008-11-19
CN101308202B CN101308202B (en) 2011-04-06

Family

ID=40027527

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2007101075109A Expired - Fee Related CN101308202B (en) 2007-05-17 2007-05-17 Parallel collection image reconstruction method and device

Country Status (2)

Country Link
US (1) US7576539B2 (en)
CN (1) CN101308202B (en)

Cited By (24)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102096055A (en) * 2010-12-14 2011-06-15 南方医科大学 Rapid and accurate reconstructing method for non-uniform sampling data of magnetic resonance imaging
CN102551723A (en) * 2012-01-16 2012-07-11 电子科技大学 Magnetic resonance parallel imaging method of multi-support vector model
CN103027681A (en) * 2011-10-06 2013-04-10 西门子公司 System used for reconstructing and parallelly obtaining mri image
CN103054580A (en) * 2011-10-21 2013-04-24 上海联影医疗科技有限公司 Rapid imaging method of blood vessel without contrast agent
CN103185878A (en) * 2011-12-27 2013-07-03 上海联影医疗科技有限公司 Magnetic resonance parallel image acquisition and image reconstruction method
CN103278784A (en) * 2013-06-02 2013-09-04 南方医科大学 Magnetic resonance parallel imaging method of multi-constraint sliding window
CN103384836A (en) * 2010-12-22 2013-11-06 皇家飞利浦电子股份有限公司 Rapid parallel reconstruction for arbitrary k-space trajectories containing GRAPPA operator
CN103389481A (en) * 2012-05-11 2013-11-13 上海联影医疗科技有限公司 Magnetic resonance frequency and phase position double-encoding sampling method and image reconstruction method
CN103728581A (en) * 2013-12-20 2014-04-16 杭州电子科技大学 SPEED fast magnetic resonance imaging method based on discrete cosine transform
CN104122520A (en) * 2013-04-26 2014-10-29 上海联影医疗科技有限公司 Magnetic resonance image reconstruction method and device
CN104166110A (en) * 2013-05-17 2014-11-26 上海联影医疗科技有限公司 Magnetic resonance parallel acquired image reconstruction method and device
CN104181481A (en) * 2013-06-28 2014-12-03 上海联影医疗科技有限公司 Magnetic resonance image acquiring and reconstructing method and apparatus
CN104360295A (en) * 2014-11-04 2015-02-18 中国科学院深圳先进技术研究院 Magnetic resonance parallel imaging method and device based on dictionary learning
CN104635188A (en) * 2013-11-12 2015-05-20 上海联影医疗科技有限公司 K-space reconstruction method and magnetic resonance imaging method
CN104698414A (en) * 2013-08-21 2015-06-10 西门子公司 Reference oversampling in sense-type magnetic resonance reconstruction
CN106597333A (en) * 2016-12-30 2017-04-26 上海联影医疗科技有限公司 Magnetic resonance parallel imaging method and magnetic resonance imaging system
CN106772167A (en) * 2016-12-01 2017-05-31 中国科学院深圳先进技术研究院 Magnetic resonance imaging method employing and device
CN107576925A (en) * 2017-08-07 2018-01-12 上海东软医疗科技有限公司 The more contrast image rebuilding methods of magnetic resonance and device
CN107656224A (en) * 2017-09-30 2018-02-02 上海联影医疗科技有限公司 A kind of MR imaging method, apparatus and system
CN108287324A (en) * 2018-01-03 2018-07-17 上海东软医疗科技有限公司 The method for reconstructing and device of the more contrast images of magnetic resonance
CN109613460A (en) * 2018-12-28 2019-04-12 上海联影医疗科技有限公司 MR imaging method, device, computer equipment and storage medium
CN111157933A (en) * 2019-01-07 2020-05-15 上海联影医疗科技有限公司 System and method for magnetic resonance image reconstruction
CN112557981A (en) * 2020-12-03 2021-03-26 川北医学院 Improved algorithm for parallel magnetic resonance imaging
CN112881957A (en) * 2017-12-26 2021-06-01 上海联影医疗科技股份有限公司 Method and system for magnetic resonance imaging

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102008045278A1 (en) * 2008-09-01 2010-03-25 Siemens Aktiengesellschaft Method for combining images and magnetic resonance apparatus
JP6261589B2 (en) 2012-08-29 2018-01-17 コーニンクレッカ フィリップス エヌ ヴェＫｏｎｉｎｋｌｉｊｋｅ Ｐｈｉｌｉｐｓ Ｎ．Ｖ． Iterative SENSE denoising using feedback
EP2924457B1 (en) * 2014-03-28 2016-11-02 Albert-Ludwigs-Universität Freiburg Half Fourier MRI with iterative reconstruction
CN112967167B (en) * 2019-12-12 2023-04-28 中国科学院深圳先进技术研究院 GPU-based image quick reconstruction method, computer-readable medium and computing device

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4678996A (en) * 1985-05-07 1987-07-07 Picker International, Inc. Magnetic resonance imaging method
US6983182B2 (en) * 1998-04-10 2006-01-03 Wisconsin Alumni Research Foundation Time resolved computed tomography angiography
GB2345139A (en) * 1998-12-24 2000-06-28 Marconi Electronic Syst Ltd MRI apparatus with continuous movement of patient
EP1454156A1 (en) * 2001-11-26 2004-09-08 Koninklijke Philips Electronics N.V. Magnetic resonance imaging method with reduced acoustic noise
US7309984B2 (en) * 2005-10-27 2007-12-18 Wisconsin Alumni Research Foundation Parallel magnetic resonance imaging method using a radial acquisition trajectory
US7397242B2 (en) * 2005-10-27 2008-07-08 Wisconsin Alumni Research Foundation Parallel magnetic resonance imaging method using a radial acquisition trajectory
CN101305908B (en) * 2007-05-17 2011-08-03 西门子公司 Signal-noise ratio computation method of parallelly collected image reconstruction

Cited By (45)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102096055A (en) * 2010-12-14 2011-06-15 南方医科大学 Rapid and accurate reconstructing method for non-uniform sampling data of magnetic resonance imaging
CN102096055B (en) * 2010-12-14 2013-01-23 南方医科大学 Rapid and accurate reconstructing method for non-uniform sampling data of magnetic resonance imaging
US9390521B2 (en) 2010-12-22 2016-07-12 Koninklijke Philips N.V. Rapid parallel reconstruction for arbitrary k-space trajectories
CN103384836A (en) * 2010-12-22 2013-11-06 皇家飞利浦电子股份有限公司 Rapid parallel reconstruction for arbitrary k-space trajectories containing GRAPPA operator
CN103027681A (en) * 2011-10-06 2013-04-10 西门子公司 System used for reconstructing and parallelly obtaining mri image
CN103027681B (en) * 2011-10-06 2016-05-04 西门子公司 For the system of the parallel MRI image obtaining of reconstruct
CN103054580A (en) * 2011-10-21 2013-04-24 上海联影医疗科技有限公司 Rapid imaging method of blood vessel without contrast agent
CN103054580B (en) * 2011-10-21 2015-07-01 上海联影医疗科技有限公司 Rapid imaging method of blood vessel without contrast agent
CN103185878A (en) * 2011-12-27 2013-07-03 上海联影医疗科技有限公司 Magnetic resonance parallel image acquisition and image reconstruction method
CN103185878B (en) * 2011-12-27 2015-04-15 上海联影医疗科技有限公司 Magnetic resonance parallel image acquisition and image reconstruction method
CN102551723B (en) * 2012-01-16 2014-01-15 电子科技大学 Magnetic resonance parallel imaging method of multi-support vector model
CN102551723A (en) * 2012-01-16 2012-07-11 电子科技大学 Magnetic resonance parallel imaging method of multi-support vector model
CN103389481A (en) * 2012-05-11 2013-11-13 上海联影医疗科技有限公司 Magnetic resonance frequency and phase position double-encoding sampling method and image reconstruction method
CN103389481B (en) * 2012-05-11 2015-09-02 上海联影医疗科技有限公司 Magnetic resonance frequency Phase Double coded sample method and image rebuilding method thereof
CN104122520A (en) * 2013-04-26 2014-10-29 上海联影医疗科技有限公司 Magnetic resonance image reconstruction method and device
CN104122520B (en) * 2013-04-26 2018-02-02 上海联影医疗科技有限公司 MR image reconstruction method and device
CN104166110A (en) * 2013-05-17 2014-11-26 上海联影医疗科技有限公司 Magnetic resonance parallel acquired image reconstruction method and device
CN104166110B (en) * 2013-05-17 2017-12-26 深圳联影医疗科技有限公司 Magnetic resonance parallel collected image reconstruction method and apparatus
CN103278784B (en) * 2013-06-02 2015-06-17 南方医科大学 Magnetic resonance parallel imaging method of multi-constraint sliding window
CN103278784A (en) * 2013-06-02 2013-09-04 南方医科大学 Magnetic resonance parallel imaging method of multi-constraint sliding window
CN106725480A (en) * 2013-06-28 2017-05-31 上海联影医疗科技有限公司 Magnetic resonance image acquisition and method for reconstructing and device
CN104181481A (en) * 2013-06-28 2014-12-03 上海联影医疗科技有限公司 Magnetic resonance image acquiring and reconstructing method and apparatus
CN104181481B (en) * 2013-06-28 2017-02-08 上海联影医疗科技有限公司 Magnetic resonance image acquiring and reconstructing method
CN104698414A (en) * 2013-08-21 2015-06-10 西门子公司 Reference oversampling in sense-type magnetic resonance reconstruction
CN104698414B (en) * 2013-08-21 2019-08-06 西门子公司 Reference over-sampling in SENSE type magnetic resonance reconstruct
CN104635188A (en) * 2013-11-12 2015-05-20 上海联影医疗科技有限公司 K-space reconstruction method and magnetic resonance imaging method
CN104635188B (en) * 2013-11-12 2018-06-19 上海联影医疗科技有限公司 K space reconstructions method and MR imaging method
CN103728581A (en) * 2013-12-20 2014-04-16 杭州电子科技大学 SPEED fast magnetic resonance imaging method based on discrete cosine transform
CN103728581B (en) * 2013-12-20 2016-01-20 杭州电子科技大学 Based on the SPEED rapid magnetic resonance imaging method of discrete cosine transform
CN104360295B (en) * 2014-11-04 2017-02-15 中国科学院深圳先进技术研究院 Magnetic resonance parallel imaging method and device based on dictionary learning
CN104360295A (en) * 2014-11-04 2015-02-18 中国科学院深圳先进技术研究院 Magnetic resonance parallel imaging method and device based on dictionary learning
CN106772167B (en) * 2016-12-01 2019-05-07 中国科学院深圳先进技术研究院 Magnetic resonance imaging method employing and device
CN106772167A (en) * 2016-12-01 2017-05-31 中国科学院深圳先进技术研究院 Magnetic resonance imaging method employing and device
CN106597333A (en) * 2016-12-30 2017-04-26 上海联影医疗科技有限公司 Magnetic resonance parallel imaging method and magnetic resonance imaging system
CN107576925A (en) * 2017-08-07 2018-01-12 上海东软医疗科技有限公司 The more contrast image rebuilding methods of magnetic resonance and device
CN107576925B (en) * 2017-08-07 2020-01-03 上海东软医疗科技有限公司 Magnetic resonance multi-contrast image reconstruction method and device
CN107656224B (en) * 2017-09-30 2020-04-21 上海联影医疗科技有限公司 Magnetic resonance imaging method, device and system
CN107656224A (en) * 2017-09-30 2018-02-02 上海联影医疗科技有限公司 A kind of MR imaging method, apparatus and system
CN112881957A (en) * 2017-12-26 2021-06-01 上海联影医疗科技股份有限公司 Method and system for magnetic resonance imaging
CN108287324B (en) * 2018-01-03 2020-05-15 上海东软医疗科技有限公司 Reconstruction method and device of magnetic resonance multi-contrast image
CN108287324A (en) * 2018-01-03 2018-07-17 上海东软医疗科技有限公司 The method for reconstructing and device of the more contrast images of magnetic resonance
CN109613460A (en) * 2018-12-28 2019-04-12 上海联影医疗科技有限公司 MR imaging method, device, computer equipment and storage medium
CN111157933A (en) * 2019-01-07 2020-05-15 上海联影医疗科技有限公司 System and method for magnetic resonance image reconstruction
CN112557981A (en) * 2020-12-03 2021-03-26 川北医学院 Improved algorithm for parallel magnetic resonance imaging
CN112557981B (en) * 2020-12-03 2023-06-13 川北医学院 Improved algorithm of parallel magnetic resonance imaging

Also Published As

Publication number Publication date
US20080285834A1 (en) 2008-11-20
CN101308202B (en) 2011-04-06
US7576539B2 (en) 2009-08-18

Similar Documents

Publication Publication Date Title
CN101308202B (en) Parallel collection image reconstruction method and device
US10139465B2 (en) Method for magnetic resonance imaging with controlled aliasing
CN1799498B (en) Quick generalized partially parallel acquisition(GRAPA) image reconstruction algorithm for magnetic resonance imaging(MRI)
CN101305908B (en) Signal-noise ratio computation method of parallelly collected image reconstruction
CN100396240C (en) MRI parallel imaging method utilizing sensitivity encoding in frequency domain
CN1530072B (en) Nuclear magnetic resonance imager and nuclear magnetic resonance imaging method
US7372269B2 (en) Magnetic resonance imaging method and apparatus
US8026720B1 (en) Rapid auto-calibrated parallel reconstruction using synthetic target coil
CN112526423B (en) Parallel magnetic resonance imaging algorithm based on conjugation and interlayer information
DE112014000556T5 (en) Metal-insensitive MR imaging
US20110184273A1 (en) System and method for moving table mri
CN106772167A (en) Magnetic resonance imaging method employing and device
US4728893A (en) Increased signal-to-noise ratio in magnetic resonance images using synthesized conjugate symmetric data
CN103890602A (en) A method and system for rapid mri acquisition using tailored signal excitation modules (rate)
Schwarz et al. GRAPPA reconstructed wave‐CAIPI MP‐RAGE at 7 tesla
Feng et al. Efficient large-array k-domain parallel MRI using channel-by-channel array reduction
JP2004524937A (en) Generalized SMASH imaging
EP2645119A1 (en) Method for rapid spin echo MRT imaging
Wech et al. High-resolution functional cardiac MR imaging using density-weighted real-time acquisition and a combination of compressed sensing and parallel imaging for image reconstruction
CN112557981B (en) Improved algorithm of parallel magnetic resonance imaging
Okanovic et al. Increasing robustness of radial GRASE acquisition for SAR-reduced brain imaging
King ASSET–parallel imaging on the GE scanner
Lazarus et al. Variable-density k-space filling curves for accelerated Magnetic Resonance Imaging
Börnert 2D-RF-pulse-encoded curved-slice imaging
Xu et al. Integrating parallel imaging with generalized series for accelerated dynamic imaging

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
ASS Succession or assignment of patent right

Owner name: SIEMENS AKTIENGESELLSCHAFT

Free format text: FORMER OWNER: SIEMENS (CHINA) CO., LTD.

Effective date: 20110209

C41 Transfer of patent application or patent right or utility model
COR Change of bibliographic data

Free format text: CORRECT: ADDRESS; FROM: 100102 NO.7, MIDDLE RING SOUTH ROAD, WANGJING, CHAOYANG DISTRICT, BEIJING TO: MUNICH, GERMANY

TA01 Transfer of patent application right

Effective date of registration: 20110209

Address after: Munich, Germany

Applicant after: Siemens AG

Address before: 100102 Beijing, Wangjing, Central South Road, No. 7, No.

Applicant before: Simens Co., Ltd. (China)

C14 Grant of patent or utility model
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20110406

Termination date: 20200517

CF01 Termination of patent right due to non-payment of annual fee